Search

Dynamic modern healthcare systems rely heavily on the contributions of computer scientists. The diagnosis process is a team effort involving many people: patients, their families, healthcare providers, researchers, and policymakers. Computer technology plays a crucial role in supporting this effort by providing a number of essential services to all of these groups. In the early stages of many diseases, a diagnosis can be made automatically using a computer-aided system, with some degree of certainty. This paper presents a hybrid optimal deep learning-based model for tuberculosis disease recognition using MRI images. Several deep learning models are combined to extract the most relevant features from MRI images. In particular, we establish a combination between vision transformer (ViTs) and Efficient-Net models in order to maximize classification accuracy. We conducted experiments to investigate the accuracy of the proposed model using the Shenzhen and Montgomery data set, and found that it yielded substantially more accurate and better results than the state of-the-art works. [ABSTRACT FROM AUTHOR]

In this paper, an adaptive neural network control problem for nonstrict-feedback nonlinear systems with an unknown backlash-like hysteresis and bounded disturbance was presented. Radial basis function neural networks (RBFNN) were used to approximate the unknown functions and the problem of the explosion of complexity problem was handled by utilizing the command filter method. Furthermore, the influence of an unknown backlash-like hysteresis input was addressed by approximating an intermediate variable. Based on the backstepping method and the command filter technique, an adaptive neural network controller was designed via the approximation abilities of RBFNN. With the help of the Lyapunov stability theory, the proposed controller ensures that all of the signals in closed-loop systems are bounded and that the tracking error fluctuates close to the origin within a bounded area. Finally, a real-world example based on the single-link manipulator was shown to demonstrate the viability of the presented approach. [ABSTRACT FROM AUTHOR]

Convolution neural networks (CNNs) have proven to be effective clinical imaging methods. This study highlighted some of the key issues within these systems. It is difficult to train these systems in a limited clinical image databases, and many publications present strategies including such learning algorithm. Furthermore, these patterns are known for making a highly reliable prognosis. In addition, normalization of volume and losses of dice have been used effectively to accelerate and stabilize the training. Furthermore, these systems are improperly regulated, resulting in more confident ratings for correct and incorrect classification, which are inaccurate and difficult to understand. This study examines the risk assessment of Fully Convolutional Neural Networks (FCNNs) for clinical image segmentation. Essential contributions have been made to this planned work: 1) dice loss and cross-entropy loss are compared on the basis of segment quality and uncertain assessment of FCNNs; 2) proposal for a group model for assurance measurement of full convolutional neural networks trained with dice loss and group normalization; And 3) the ability of the measured FCNs to evaluate the segment quality of the structures and to identify test examples outside the distribution. To evaluate the study’s contributions, it conducted a series of tests in three clinical image division applications such as heart, brain and prostate. The findings of the study provide significant insights into the predictive ambiguity assessment and a practical strategies for outside-distribution identification and reliable measurement in the clinical image segmentation. The approaches presented in this research significantly enhance the reliability and accuracy rating of CNNbased clinical imaging methods. [ABSTRACT FROM AUTHOR]

Over the last few years, the amount of available information has increased exponentially in all professional fields, including the medical field. Modern-day patients have access to a wealth of medical information, whether it be from brochures, newspapers, television campaigns, or internet documents. To facilitate and accelerate the search for medical information, more precise systems have been implemented, such as visual question-and-answer systems. A visual question-and-answer system is designed to provide direct and precise answers to questions asked in natural language. In this context, we propose an optimal deep neural network model based on an adaptive optimization algorithm, which takes medical images and natural language questions as input, then provides precise answers as output. Our model starts by classifying medical questions following an embedding phase. We then use a deep learning model for visual and textual feature extraction and emergence. In this paper, we aim to maximize the accuracy rate and minimize the number of epochs in order to accelerate the process. This is a multi-objective optimization problem. The selection of deep learning model parameters, such as epoch number and batch size, is an essential step in improving the model, thus, we use an adaptive genetic algorithm to determine the optimal deep learning parameters. Finally, we propose a dense layer for answer retrieval. To evaluate our model, we used the ImageCLEF 2019 VQA data set. Our model outperforms existing visual question-and-answer systems and offers a significantly higher retrieval accuracy rate. [ABSTRACT FROM AUTHOR]